# Welcome to the 'Introduction to GAM & GAMM' course

The course material is accessible via the menu on the left.

Commencing with a basic introduction to generalised additive models (GAM) to analyse continuous data, count data and binary/proportional data. In the second part of the course generalised additive mixed effects models (GAMM) are introduced to analyse nested data.

During the course several case studies are presented, in which the statistical theory is integrated with applied analyses in a clear and understandable manner.

#### Keywords

Introduction to GAM. Revision of mixed effects models. Poisson, negative binomial, binomial GAM. Overdispersion. lmer. GAM and GAMM in mgcv and gamm4.

#### Schedule

Monday & Tuesday

• Revision linear regression exercise
• Introduction to GAM using the gam function from mgcv. This module is based on Chapter 2 in Zuur (2012).
• Three exercises (Gaussian, Poisson, negative binomial and Bernoulli GAM) using the mgcv package in R.

Wednesday

• Revision mixed effects models for 1-way nested data using nlme and lme4
• Random intercept models.
• Sketching fitted values.
• One exercise.
• Based on Chapter 4 in Zuur et al. (2013).

Wednesday - Friday

• Introduction to GAMM for nested count data, binary data and proportional data (Poisson, binomial) using gamm and gamm4.
• Based on various chapters in Zuur et al. (2014).
• Five exercises.
• Technical details of smoothers. This module is based on Chapter 3 in Zuur (2012).
• What to write in a paper.
• Time allowing: Two-dimensional smoothers
• Time allowing: Pointers for zero inflation, temporal correlation, spatial correlation.

#### Course material

Various chapters from:

• A Beginner’s Guide to GAM with R. (2012). Zuur.
• A Beginner’s Guide to GLM and GLMM with R. (2013). Zuur, Hilbe, Ieno.
• A Beginner’s Guide to GAMM with R (2014). Zuur, Saveliev, Ieno.

The above three books are not part of the course fee and are available from www.highstat.com. The course can be attended without purchasing these books. All R solutions codes and pdf files of powerpoint files are provided before the start of the course.

This is a non-technical course. You need to bring your own laptop.

#### Pre-required knowledge

Fluent knowledge of R, data exploration, multiple   linear regression & GLM (Poisson, binomial,  negative binomial). Working knowledge of mixed modelling.

#### Cancellation Policy

What if you are not able to participate? Once participants are given access to course exercises with R solution codes, pdf files of book chapters, pdf files of powerpoint files and video solution files, all course fees are non-refundable. However, depending for the circumstances for cancellation we may be able to offer you the option to attend a future course or you can authorise a colleague to attend this course.